A variety of different applications use sensor systems to detect the movement of an underlying object. Sensors employing microelectromechanical systems (MEMS) devices are increasingly used in such applications due to their relatively small size and their capability to detect relatively small changes in the measured item. MEMS devices, such as an accelerometer, typically employ a movable, inertial mass formed with one or more fixed, non-moving structures. For example, in a MEMS accelerometer, the inertial mass may be suspended in a plane above a substrate and movable with respect to the substrate. The movable structure and the fixed structures form a capacitor having a capacitance that changes when the movable structure moves relative to the fixed structures in response to applied forces, such as along a predetermined axis of the device, e.g., x-, y- and z-axes.
Currently, accelerometers may be used for monitoring the interaction of users with electronic devices, such as gaming devices, cell phones, personal digital assistants, etc. The accelerometer in the devices may detect a movement above a particular threshold or a change in orientation. However, accelerometers experience a constant 1 g field due to the earth's gravitational field. This 1 g field is in an arbitrary direction depending upon the orientation of the device. In order to find the actual acceleration of an object (relative to the earth), a 1 g threshold is typically set in the device in order to eliminate the contribution from the earth's gravitational field. This means, however, that a device typically can only detection acceleration greater than 1 g. High pass filters may be used in an attempt to compensate for this shortcoming, but this solution does not detect extremely slow interactions. In some cases, however, users move electronic devices in a slow manner and do not cause greater than a 1 g change. Other problems with monitoring the interaction of users with electronic devices is determining when users stop moving the device. The device may not be placed in a known orientation so detecting inactivity has some similar challenges as detecting movement or activity of the device.